Dendritic cells are potent antigen presenting cells capable of activating quiescent lymphocytes. Thus, these cells may be instrumental in the in vivo generation of tumor reactive T cells. We have studied several methods to gene modify, activate, and induce the proliferation of dendritic cells in an effort to increase their effectiveness in stimulating an anti-tumor response. The stable introduction of antigen genes into dendritic cells may be a more effective antigen loading method compared to peptide pulsing, since the antigen would be constitutively expressed in vivo and could express multiple epitopes from a single gene. We have developed a technique to stably gene modify primary murine and human dendritic cells using retroviral vectors. Because retroviral transduction requires proliferating cells, we gene modified dividing hematopoietic progenitor cells followed by in vitro differentiation into dendritic cells. We have demonstrated that murine dendritic cells transduced with a model antigen gene can treat established pulmonary metastases. In addition, we have transduced human dendritic cells with the melanoma antigens MART-1 and GP100 and have demonstrated that these cells can be used to generate tumor-specific T- cells from peripheral blood lymphocytes. T-cell cultures generated from GP100 transduced DCs could recognize at least 3 distinct epitopes from GP100, thereby demonstrating one of the advantages of gene transduction compared to pulsing with a single peptide epitope. Most recently, we have demonstrated that lentiviral vectors can gene modify non- proliferating monocyte-derived dendritic cells. Because dendritic cells are rare cell types in vivo and difficult to grow in large numbers, we are studying methods of in vivo priming with Flt3L, a recently discovered cytokine that has been shown to increase the number of circulating dendritic cells in murine models as well as normal volunteers. We are currently conducting a clinical trial using Flt3L in combination with melanoma peptide vaccination. We have found that subcutaneous administration of Flt3L increased circulating cells with a dendritic cell phenotype capable of stimulating T cells presumably through the increased expression of costimulatory molecules. This ability to mobilize cells with a DC phenotype which are capable of stimulating greater T cell proliferation may translate into a stronger in vivo anti-tumor response. The activation state of dendritic cells may be critical in determining their ability to effectively stimulate T-cells. Therefore, we are studying methods to fully activate dendritic cells, such as upregulating costimulatory molecules with CD40L stimulation, to determine if activated dendritic cells can more effectively generate an antitumor response. We have found that dendritic cells can be optimally stimulated using a combination of activation signals, such as CD40L and LPS. By utilizing gene transfer into hematopoietic cells, we have also developed a novel method to generalize immunotherapy to common cancers, such as ovarian, breast, and colon cancer. We have developed chimeric antibody/T-cell receptor genes which combine variable regions from mAb with T-cell signaling chains. We have demonstrated that the insertion of these chimeric receptor genes into T-cells allows them to recognize tumor antigens defined by the receptors. We are currently conducting a clinical trial for ovarian cancer patients based on this strategy. In addition, we have transduced hematopoietic stem cells with chimeric receptor genes, and have demonstrated that mice reconstituted with the gene-modified bone marrow cells exhibit decreased tumor growth when challenged with tumor cells expressing the corresponding antigens. Transducing hematopoietic stem cells with chimeric receptor genes may allow the continuous in vivo production of lymphocytes, macrophages, neutrophils, and NK cells directed against the tumor. Most recently, we have designed chimeric receptors that recognize KDR, the receptor for VEGF that is overexpressed on tumor vasculature. This approach, which would be broadly applicable to many types of cancer, may allow us to target tumor vasculature using transduced T cells. - gene therapy, immunotherapy, dendritic cells, - Human Subjects